EGFR activation mutations are frequently present in non-smoking related lung cancer. Most of these tumors are addicted to EGFR signaling and can be effectively treated with tyrosine kinase inhibitors (TKI), such as gefitinib. High sensitivity to gefitinib in lung cancer is closely correlated with dependence on AKT activation in response to EGFR signaling, and gefitinib treatment suppresses AKT activation. The restoration of AKT function was recently found to be a key step in EGFR mutant lung tumors that acquired resistance to TKI, and one potential mechanism involves the amplification of c-MET which bypasses the requirement for EGFR, rendering EGFR inhibitors ineffective. LKB1 is commonly known as a tumor suppressor because somatic LKB1 inactivation mutation is one of the most frequently mutated genes in smoking-related lung cancer. However, LKB1 mutations were rarely found in lung cancers with EGFR mutations. We have discovered a novel, potentially oncogenic role for LKB1 in lung cancers that are addicted to EGFR signaling. We found the suppression of LKB1 expression led to apoptosis in six cell lines in which either EGFR or AKT is constitutively active, but not in two cell lines without EGFR or AKT activation. More importantly, EGFR mutant cells with acquired TKI-resistance through two known mechanisms are still prone to apoptosis induced by LKB1 depletion. Mechanistically, we found that EGFR/AKT activation led to increased phosphorylation of FoxO3A at threonine 32 (Thr32) in LKB1 wild-type cells, but not in LKB1-null cells. Depletion of LKB1 in the cells with wild- type LKB1 resulted in attenuation of that phosphorylation of FoxO3A by activated EGFR/AKT, while the restoration of LKB1 function in LKB1-null cells re-established EGFR/AKT mediated FoxO3A phosphorylation. Upon expanding our analysis to other AKT targets, using three different isogenic LKB1 knockdown cell line pairs and a phospho-specific antibody microarray, we observed that there was a requirement for LKB1 in the phosphorylation of other AKT down-stream targets, including BAD (Ser136), FoxO1 (Ser319), FoxO4 (Ser197) and GSK32 (Ser9). Because the phosphorylation of these sites by AKT suppresses apoptosis, the requirement of LKB1 suggests that LKB1 may have a pro-apoptotic role in tumor cells that are addicted to EGFR signaling. In summary, we believe LKB1 plays dual roles in lung cancer development. While it is a tumor-suppressor in lung cancer caused by smoking, we hypothesize that the integrity of the LKB1/AMPK pathway is a critical determinant of the phosphorylation and inactivation of pro-apoptotic proteins in non-smoking related lung cancers that are addicted to EGFR signaling. A better understanding of this novel, potential pro-oncogenic role of the LKB1/AMPK pathway in NSCLC and its role in chemosensitivity may have a direct impact on the targeted clinical use of existing therapies, such as EGFR inhibitors, and may provide a molecular basis for future implementation of personalized therapy.